基于数值模拟的注塑成形工艺参数优化方法研究
|
摘要
在注塑成形过程中,熔体在型腔内的状态与变化直接影响产品最终的质量和性能。运用数值模拟软件模拟成形过程,可以预测成形后的产品可能出现的缺陷,从而辅助模具设计和成形工艺设置,成为提高产品质量的重要手段。虽然运用数值模拟软件能够代替试模来反复验证工艺参数是否合适,但却难以得到最优工艺参数。
本文应用Moldflow软件按照正交表进行数值模拟分析,获取了该工艺参数下的制品质量,预测了制品翘曲、收缩率、熔接痕等缺陷;通过选取并分析企业实际生产中产生欠注的某塑料制品,从工艺上和结构上提出了三种解决方案,并成功解决了该问题。基于CAE软件模拟结果图,量化翘曲结果为数值形式。选用收缩率在制品内的差值(最大值和最小值的差值)和翘曲作为考察质量指标,选用工艺参数熔体温度、模具温度、注射时间、保压时间和保压压力为影响因素,应用Taguchi DOE技术研究了工艺参数对制品质量的影响,确定了工艺参数对产品质量指标的影响程度,获取近似最优工艺参数组合,并将该组合在CAE软件中进行验证,结果显示了正交实验设计在理论上可行性。以数值模拟获取的数据为样本,采用支持向量机建立注塑产品质量和工艺参数之间关系的模型,代替了用CAE软件进行模拟,使得制品质量预测更加快捷,并基于该关系模型,采用遗传算法获得了成形工艺的全局最优解。
It has a direct impact on the quality and performances of products in the injection molding process include the state of the melt and the cavity. The use of numerical simulation software to simulate forming process can predict shaped product's defect as well as aid mold design and molding process settings, which is an important tool to improve product quality. Although the use of numerical simulation software can replace the test mode to validate the suitability of process parameters without trying again and again, it is difficult to obtain optimal process parameters.
In this paper, by using Moldflow software with the orthogonal, forming process has simulated, then the quality of products on some special process parameters is obtained and some defect such as warpage, shrinkage and weld line are well predicted; Considering the problems existing on the plastic products by analyzing insufficient injection products, three candidates solutions inspired from process and structure to solve those problem is proposed. A numerical result is quantified with warpage numerical simulation results getting from CAE simulation software. Choosing Shrinkage rate in the margin (the difference between maximum and minimum values) and warping as the quality inspection factors and the process parameters such as melt temperature, mold temperature, injection time, packing time and packing pressure as the influence factors, the impact on product quality of process parameters is investigated by applying the technique of Taguchi DOE, then the influence degree of the process parameters on product quality and the approximate optimal combination of process parameters are obtained. Using those parameters based on CAE software, the results strongly demonstrates the feasibility of the orthogonal experimental design in the theory. Taking numerical simulation data as samples, the relationship model between injection molding process parameters and product quality is established by suing Support Vector Machine, which can replace the CAE software to simulate and make the prediction of products quality more advantageous. Based on that relational model, the global optimal solution of forming process is obtained by applying genetic algorithms
本文应用Moldflow软件按照正交表进行数值模拟分析,获取了该工艺参数下的制品质量,预测了制品翘曲、收缩率、熔接痕等缺陷;通过选取并分析企业实际生产中产生欠注的某塑料制品,从工艺上和结构上提出了三种解决方案,并成功解决了该问题。基于CAE软件模拟结果图,量化翘曲结果为数值形式。选用收缩率在制品内的差值(最大值和最小值的差值)和翘曲作为考察质量指标,选用工艺参数熔体温度、模具温度、注射时间、保压时间和保压压力为影响因素,应用Taguchi DOE技术研究了工艺参数对制品质量的影响,确定了工艺参数对产品质量指标的影响程度,获取近似最优工艺参数组合,并将该组合在CAE软件中进行验证,结果显示了正交实验设计在理论上可行性。以数值模拟获取的数据为样本,采用支持向量机建立注塑产品质量和工艺参数之间关系的模型,代替了用CAE软件进行模拟,使得制品质量预测更加快捷,并基于该关系模型,采用遗传算法获得了成形工艺的全局最优解。
It has a direct impact on the quality and performances of products in the injection molding process include the state of the melt and the cavity. The use of numerical simulation software to simulate forming process can predict shaped product's defect as well as aid mold design and molding process settings, which is an important tool to improve product quality. Although the use of numerical simulation software can replace the test mode to validate the suitability of process parameters without trying again and again, it is difficult to obtain optimal process parameters.
In this paper, by using Moldflow software with the orthogonal, forming process has simulated, then the quality of products on some special process parameters is obtained and some defect such as warpage, shrinkage and weld line are well predicted; Considering the problems existing on the plastic products by analyzing insufficient injection products, three candidates solutions inspired from process and structure to solve those problem is proposed. A numerical result is quantified with warpage numerical simulation results getting from CAE simulation software. Choosing Shrinkage rate in the margin (the difference between maximum and minimum values) and warping as the quality inspection factors and the process parameters such as melt temperature, mold temperature, injection time, packing time and packing pressure as the influence factors, the impact on product quality of process parameters is investigated by applying the technique of Taguchi DOE, then the influence degree of the process parameters on product quality and the approximate optimal combination of process parameters are obtained. Using those parameters based on CAE software, the results strongly demonstrates the feasibility of the orthogonal experimental design in the theory. Taking numerical simulation data as samples, the relationship model between injection molding process parameters and product quality is established by suing Support Vector Machine, which can replace the CAE software to simulate and make the prediction of products quality more advantageous. Based on that relational model, the global optimal solution of forming process is obtained by applying genetic algorithms
引文
[1]路书芬,注塑成形工艺参数对制品质量影响的实验研究,郑州大学硕士学位论文,2006.4
[2]许文耀,注塑成形中注射熔体温度的研究,华南理工硕士论文,2003
[3]王蓓,王利霞,李倩等,注塑成形工艺控制技术及研究进展,工程塑料应用,2003,31(5):31-34
[4]王利霞,基于数值模拟的注塑成形工艺优化及制品质量控制研究,郑州大学博士论文,2004
[5]T.Pertova and D.Kazmer,In exportation of Phenomenological Models in a Hybrid Neural Network for Quality Control of Injection Molding,Polym.-Plast.Technol.Eng.,1999,38(1):1-18
[6]Qin Zou and Gbariel Ari,A Total System Approach to Improving an Injection Molded Plastic Part Quality,SPE ANTEC,1994:353-359
[7]J.C.Rowlnad and K.Ho-Le,Process Quality Assurance for Injection Molding of Thermoplastic Polymers,SPEANTEC,1994:342-347
[8]M.Sereer,I.Catie,G.Barie and D.Pekrovie,Trend Regulation of Injection Molding Process,SPEANTEC,1997:556-560
[9]Hua Ye,Optimization of injection molding Process with genetic algorithms,ANTEC,1999:594-597
[10]Hua Ye,Y'mghui Wu,K.K.Wnag,An optimization scheme for Part quality in injection molding,MD-vol,CAE and Intelligent Processing of Polymeric Materials,ASME,1997:139-149
[11]Hua Ye,Phillip M.Leopold,Minimizing Part sink Marks using C-Mold and genetic optimization algorithm,Antec,1999:589-593
[12]王德翔,刘来英等,基于人工神经网络技术的注塑成形工艺参数优化,模具技术,2001,6:1-4郑州大学硕士学位论文
[13]G.H.Choi,K.D.Lee,N.Chnag,S.GKim,optimization of process parameters of Injection Molding with Neural Network Application in a Process Simulation Environment,Annals of the CIRP,1994,43(1):449-452
[14]S.Liu,J.Wu,J.Chnag,An experimental matrix design to optimize the weld line strength in injection molded Parts,Polymer Engineering and Science,2001,40(5):1256-1262
[15]S.C.Malgnmaera,A.Mnaisali,Effects of processing parameters on the tensile Properties of weld lines in injection molded thermo Pasties,Polymer Engineering and Science,1981,21(10):586-593
[16]S.C.Malgumaera,A.I.Mnaisali,D.C.Riggs,Weld line structure and properties in injection molded Polypropylene,Polymer Engineering and Science,1981,21(17):1149-1155
[17]R.Selden,Eeffet of processing on weld line strength in five thermoplastics,Polymer Engineering and Science,1997,37(1):205-218
[18]S.Picearolo,M.Saiu,Knit-line structure the injection molding of Polystyrene,Plastics and Rubber Processing and Applications,1988,10(1):11-16
[19]S.Hashemi,Influence of temperature on weld line strength of injection molded short glass fiber styrene anhydride Polymer composites,Plastics,Rubber and Composites,2002,31(7):318-324
[20J E.Worden,S.Kushion,Weld line integrity of PCA/BS engineering blends,ANTEC,1991:2653-2655
[21]G.Titomnalio,S.Pieemalo,Rallis,Knit lines structure in injection molding of nylon6:Relation to first solidified sample cross section,Polymer Engineering and Science,1989,29(4):209-213
[22]王利霞,张娜等,应用Taguchi实验设计法最小化注塑成形制品沉降斑,塑料工业,2004,32(4):26-28
[23]王仁庆译,塑料注塑工艺,成都:四川科学技术出版社,1988
[24]蒋华伟,基于ACE软件的注塑工艺研究,吉林工业大学硕士学位论文,2000
[25]刘来英主编,注塑成形工艺,机械工业出版社,2004
[26]王兴天主编,注塑技术与注塑机,化学工业出版社,2005郑州大学硕士学位论文
[27]肖长江,刘春太等,注塑制件熔接线的形成、性能和预测,工程塑料应用,2003,31(3):17-20
[28]陆平昌,无定形聚合物注塑制品中熔接缝强度的研究,华南理工大学硕士学位论文,2004
[29]于同敏,路全胜等,注塑成形收缩与塑件尺寸精度,模具工业,1995,8:36-41
[30]张治华,塑料收缩性,中国石化出版社,1999
[31]王庆利,基于Moldflow注塑制品收缩规律模拟实验研究,大连理工大学硕士学位论文,2005
[31]牛克中,成形收缩率塑料及其在生产中的控制方法,河南机电高等专科学校学报,2000,3:28-32
[32]姜勇道,李德群,华塑CAE3D软件翘曲分析及应用,模具工程,2005,3:44-46
[33]杨凤霞,注塑制品常见表观缺陷产生原因及其对策,塑料,2003,3(26):59-63
[34]姜勇道,李德群,华塑CAE3D软件翘曲分析及应用,模具工程,2005,3:44-46
[35]Rafael C.Gonzalez,Richard E.Woods.Digital Image Processing(Second Edition)[M].Library of Congress Cataloging-in-Publication,2007.8
[36]贺华波,李红林,刘军,等.Moldflow在对讲机天线零件注塑成形中的应用[J].轻工机械,2005,23(1):29-31
[37]王爱玲,叶明生,邓秋香编著.MATLBAR2007图像处理技术与应用[M].电子工业出版社,2008.1
[38]99.Tatiana Petrova,David Kazmer,Development of a hybrid Neural network for quality control of Injection molding,ANTEC'98,665-659
[39]郑少华,姜奉华编著,试验设计与数据处理,中国建材工业出版社,2004.3
[40]栾军,现代实验设计优化方法,上海交通大学出版社,1995
[41]张学工译.统计学习理论的本质.北京:清华大学出版社,2000
[42]邓乃扬,田英杰著.数据挖掘中的新方法--支持向量机.北京:科学出版社,2004
[43]B Schoclkopf,A Smola.Learning with Kernels[M].Cambridge.MA:MIT Press,2002.
[44]Wartell Zachary Justin.Stereoscopic head-tracked displays:Analysis and development of display algorithms[D].Georgia Institute of Technology,2001.
[45]Chen C R,Ramaswamy H S,Modeling and optimization of variable retort temperature(VRT) thermal processing using coupled neural networks and genetic algorithms[J],Journal of Food Engineering,2002,53(3):209-220.
[46]Christopher R houck,Jeffery A Joines,Michael G Kay,A Genetic Algorithm for Function Optimization:A Matlab Implementation[BD],www.ie.ncsu.edu/mirage,1996
[47]Blanco A,Delgado M,Pegalajar M C,A real-coded genetic algorithm for training recurrent neural networks[J],Neural Networks,2001,14(1):93-105
[2]许文耀,注塑成形中注射熔体温度的研究,华南理工硕士论文,2003
[3]王蓓,王利霞,李倩等,注塑成形工艺控制技术及研究进展,工程塑料应用,2003,31(5):31-34
[4]王利霞,基于数值模拟的注塑成形工艺优化及制品质量控制研究,郑州大学博士论文,2004
[5]T.Pertova and D.Kazmer,In exportation of Phenomenological Models in a Hybrid Neural Network for Quality Control of Injection Molding,Polym.-Plast.Technol.Eng.,1999,38(1):1-18
[6]Qin Zou and Gbariel Ari,A Total System Approach to Improving an Injection Molded Plastic Part Quality,SPE ANTEC,1994:353-359
[7]J.C.Rowlnad and K.Ho-Le,Process Quality Assurance for Injection Molding of Thermoplastic Polymers,SPEANTEC,1994:342-347
[8]M.Sereer,I.Catie,G.Barie and D.Pekrovie,Trend Regulation of Injection Molding Process,SPEANTEC,1997:556-560
[9]Hua Ye,Optimization of injection molding Process with genetic algorithms,ANTEC,1999:594-597
[10]Hua Ye,Y'mghui Wu,K.K.Wnag,An optimization scheme for Part quality in injection molding,MD-vol,CAE and Intelligent Processing of Polymeric Materials,ASME,1997:139-149
[11]Hua Ye,Phillip M.Leopold,Minimizing Part sink Marks using C-Mold and genetic optimization algorithm,Antec,1999:589-593
[12]王德翔,刘来英等,基于人工神经网络技术的注塑成形工艺参数优化,模具技术,2001,6:1-4郑州大学硕士学位论文
[13]G.H.Choi,K.D.Lee,N.Chnag,S.GKim,optimization of process parameters of Injection Molding with Neural Network Application in a Process Simulation Environment,Annals of the CIRP,1994,43(1):449-452
[14]S.Liu,J.Wu,J.Chnag,An experimental matrix design to optimize the weld line strength in injection molded Parts,Polymer Engineering and Science,2001,40(5):1256-1262
[15]S.C.Malgnmaera,A.Mnaisali,Effects of processing parameters on the tensile Properties of weld lines in injection molded thermo Pasties,Polymer Engineering and Science,1981,21(10):586-593
[16]S.C.Malgumaera,A.I.Mnaisali,D.C.Riggs,Weld line structure and properties in injection molded Polypropylene,Polymer Engineering and Science,1981,21(17):1149-1155
[17]R.Selden,Eeffet of processing on weld line strength in five thermoplastics,Polymer Engineering and Science,1997,37(1):205-218
[18]S.Picearolo,M.Saiu,Knit-line structure the injection molding of Polystyrene,Plastics and Rubber Processing and Applications,1988,10(1):11-16
[19]S.Hashemi,Influence of temperature on weld line strength of injection molded short glass fiber styrene anhydride Polymer composites,Plastics,Rubber and Composites,2002,31(7):318-324
[20J E.Worden,S.Kushion,Weld line integrity of PCA/BS engineering blends,ANTEC,1991:2653-2655
[21]G.Titomnalio,S.Pieemalo,Rallis,Knit lines structure in injection molding of nylon6:Relation to first solidified sample cross section,Polymer Engineering and Science,1989,29(4):209-213
[22]王利霞,张娜等,应用Taguchi实验设计法最小化注塑成形制品沉降斑,塑料工业,2004,32(4):26-28
[23]王仁庆译,塑料注塑工艺,成都:四川科学技术出版社,1988
[24]蒋华伟,基于ACE软件的注塑工艺研究,吉林工业大学硕士学位论文,2000
[25]刘来英主编,注塑成形工艺,机械工业出版社,2004
[26]王兴天主编,注塑技术与注塑机,化学工业出版社,2005郑州大学硕士学位论文
[27]肖长江,刘春太等,注塑制件熔接线的形成、性能和预测,工程塑料应用,2003,31(3):17-20
[28]陆平昌,无定形聚合物注塑制品中熔接缝强度的研究,华南理工大学硕士学位论文,2004
[29]于同敏,路全胜等,注塑成形收缩与塑件尺寸精度,模具工业,1995,8:36-41
[30]张治华,塑料收缩性,中国石化出版社,1999
[31]王庆利,基于Moldflow注塑制品收缩规律模拟实验研究,大连理工大学硕士学位论文,2005
[31]牛克中,成形收缩率塑料及其在生产中的控制方法,河南机电高等专科学校学报,2000,3:28-32
[32]姜勇道,李德群,华塑CAE3D软件翘曲分析及应用,模具工程,2005,3:44-46
[33]杨凤霞,注塑制品常见表观缺陷产生原因及其对策,塑料,2003,3(26):59-63
[34]姜勇道,李德群,华塑CAE3D软件翘曲分析及应用,模具工程,2005,3:44-46
[35]Rafael C.Gonzalez,Richard E.Woods.Digital Image Processing(Second Edition)[M].Library of Congress Cataloging-in-Publication,2007.8
[36]贺华波,李红林,刘军,等.Moldflow在对讲机天线零件注塑成形中的应用[J].轻工机械,2005,23(1):29-31
[37]王爱玲,叶明生,邓秋香编著.MATLBAR2007图像处理技术与应用[M].电子工业出版社,2008.1
[38]99.Tatiana Petrova,David Kazmer,Development of a hybrid Neural network for quality control of Injection molding,ANTEC'98,665-659
[39]郑少华,姜奉华编著,试验设计与数据处理,中国建材工业出版社,2004.3
[40]栾军,现代实验设计优化方法,上海交通大学出版社,1995
[41]张学工译.统计学习理论的本质.北京:清华大学出版社,2000
[42]邓乃扬,田英杰著.数据挖掘中的新方法--支持向量机.北京:科学出版社,2004
[43]B Schoclkopf,A Smola.Learning with Kernels[M].Cambridge.MA:MIT Press,2002.
[44]Wartell Zachary Justin.Stereoscopic head-tracked displays:Analysis and development of display algorithms[D].Georgia Institute of Technology,2001.
[45]Chen C R,Ramaswamy H S,Modeling and optimization of variable retort temperature(VRT) thermal processing using coupled neural networks and genetic algorithms[J],Journal of Food Engineering,2002,53(3):209-220.
[46]Christopher R houck,Jeffery A Joines,Michael G Kay,A Genetic Algorithm for Function Optimization:A Matlab Implementation[BD],www.ie.ncsu.edu/mirage,1996
[47]Blanco A,Delgado M,Pegalajar M C,A real-coded genetic algorithm for training recurrent neural networks[J],Neural Networks,2001,14(1):93-105